Dependence on heroin and prescription pain relievers has almost tripled since 2000, resulting in dramatic increases in opioid dependence treatment admissions. However, among agonist-maintained patients continued drug use is common and associated with high rates of relapse and treatment drop-out. Although counseling has been shown to be effective, it is costly and some patients dislike counseling, others have responsibilities that make attendance difficult, and patients in rural settings often have limited access to psychotherapy. Thus, there is a clear need to develop additional acceptable and cost-effective treatments. Interactive Voice Response (IVR) systems, which have been shown to effectively augment brief interventions for substance abuse, are automated, computer-based systems delivered via phone and use voice or keys to access different menus. IVR systems can be accessed from any phone rather than only specified technology (e.g., smart phones), and offer advantages of low cost, consistent delivery, expanded access, and 24-hour availability of immediate therapeutic intervention. The Recovery Line is a Cognitive Behavioral Therapy (CBT)-based IVR system to reduce substance use in patients receiving opioid agonist maintenance. We recently completed a pilot randomized 4-week trial which showed significant reduction in cocaine use and increased coping skill efficacy, but patients called less time than expected, suggesting methods to improve patient use may further improve efficacy. This Stage Ib application proposes three phases to develop system functions to increase patient system use and to test those functions. Specific aim 1 (Phase 1) is to develop and evaluate customized therapeutic recommendations based on coping skills assessments. Patients will be directed to modules to help them with assessed coping deficiencies based on their answers to brief preliminary questions. Specific aim 2 (Phase 2) is to develop and test system reminder messages to evaluate optimal latency (time since missed call) and message content. We will evaluate gain-framed and loss-framed messages because they have been effective in message-based smoking interventions. We will evaluate reminder message latency because short reminder latencies may be effective initially but frequent reminders may become aversive and ineffective over time. Phases 1 and 2 features will be incorporated into the final Recovery Line prototype. Specific aim 3 (Phase 3), is to conduct a 12-week pilot randomized (N=60), clinical trial with a 3 month post-treatment follow- up to obtain data regarding the feasibility, acceptability and efficacy of the developed Recovery Line (compared to TAU). Efficacy will be evaluated for the two primary outcome domains of the proportion of urine screens negative for illicit drugs and monthly days of illicit drug abstinence. Secondary outcome measures will be retention in treatment and coping skills efficacy.